The Sensory Motor Performance Program specializes in the study of human motor control in normal and impaired subjects. Current projects examine upper limb, lower limb, neck, and eye movements using a combination of experimental and computer-based analysis. This proposal requests a graphics supercomputer to enhance our experimental and computational capabilities, and to link our research teams via a high speed-computer network. This advanced graphics technology will make possible the development of three-dimensional computer graphics models of the neuromusculoskeletal system, which will significantly enhance our studies of movement. Specifically, Dr. Delp will use the graphics supercomputer to develop interactive computer models of individual subjects with cerebral palsy. These biomechanical models will be constructed from MRI data to simulate gait abnormalities and proposed surgical corrections. Drs. Peterson, Keshner, and Delp will use the graphics supercomputer to develop anatomically accurate models of the head and neck musculature as an extension of their studies of human and primate neck movement. Dr. Buchanan's investigations of arm movement will also be greatly enhanced through the development of a ~virtual arm~ to study of the role played by dynamic forces in trajectory planning. The computational and graphics performance of the proposed computer will result in a more detailed and accurate arm model and make it feasible to perform real-time dynamic simulations. Drs. Peterson, Kettner et al. will exploit the computational power of the supercomputer to simulate cerebellar control of eye movements. They have developed a model that simulates the behavior of the dense neural networks of the cerebellum, which contains 10(10) neurons, and is currently represented with a network containing 6000 simulated neurons. A single simulation currently takes a day or more on a fast desktop workstation. Further advances will require more complex networks and performing multiple simulations. The proposed supercomputer will make this research much more tractable. Dr. Rymer~s studies of spasticity and fatigue aim to quantify how the mechanical properties of muscle are altered by necrologic disease and fatigue. These studies rely on accurate modeling and simulation of limb motion, limb impedance, and motor control, and would be greatly enhanced by the advanced computational capacity of the proposed computer. The unsurpassed capabilities of the proposed supercomputer will enable a diverse team of NIH-sponsored scientists pursue promising new directions in motor control research.